Stuck trying to figure out how to parse a file that we get from fedwire the general format is in attached myfile.txt There can be any number of records. Within the records the data elements are separated by a tag enclosed in {} which represent a field name, the data following the tag is associated with that field, up until we reach next tag {}. Each record can have different tags and different number of tags, as in the attached example all three records have the same tags {1100}, {1110},{1120},{1510},{1520},{2000},{3100},{3320},{3400},and {3600}. However you can see record three then jumps to {4200}, whereas records 1and 2 the next tag is {3620}.
I have tried using input @, but struggling with that, stopping after field2, also some of the data within the tag can be variable length, so can't always set the field value to extract exact length of the data, so trying to figure out how to parse by tags {} and capture data.
options ls=max;
filename fedwire "%sysfunc(pathname(WORK))\fedwire.txt";
data have;
input;
file fedwire;
put _infile_;
datalines;
{1100}30P N{1110}08030815FT03{1120}20210803L1B7832F00013308030815FT03{1510}1000{1520}20210801B1Q8161C001707{2000}002020230000{3100}021000018BANKOBOB*{3320}F9S2108035631900*{3400}121125631AAA BBB CCCC DD EE*{3600}CTP{3620}3*FA6FE938-A6F6-4EE9-A062-6B87FC752339*{3700}SUSD0,00*{3710}USD203000,*{4200}D1602204277*ALOT OF WRITTING STUFF 98203*{5000}D02628651*MR WARREN JOHN BLAHALBH*FLAT 30 135 AAAAAAA WALK*LONDON N1 7RR GB*{5100}D8033779245*COOTTS AND COMP'L. KING DIV*73 MAN STR, EAST HSE, 3RD FL*D*{6000}HELEN BOBS STUFF*ROUTE: 122228444*
{1100}30P N{1110}08031116FT03{1120}20210803L1B7832F00031008031116FT03{1510}1000{1520}20210803B1QGC06C006237{2000}000300899500{3100}021000021NONEOFBUS*{3320}98566273215FS*{3400}121222484AAA BBB CCCC DD EE*{3600}CTP{3620}3*9A294483-5377-4A00-9362-8C95FEF4229F*{3700}SUSD5,00*{3710}USD6000,00*{4100}F124448405*MOREBLAH BLAH BLAH STUFF*{4200}D88040660033579*EDUARDO BLAH BLAH HERNANDEZ, E*STADOS UNIDOS,11111 LILAC ,TEXAS*{4320}SWF OF 21/09/03*{5000}DCLIENTE:13770683*INDUSTY S A DE C V*BRALES C P*45694 EL SALTO JALISCO MEXICO*{5100}BRGIOMFE33XMT*
{1100}30P S{1110}08031811FT03{1120}20210803L1B7832F00101008031811FT03{1510}1631{1520}20210803QMGAT015001875{2000}000440962083{3100}125108405CANTSEE ME*{3320}F001BB*{3400}121045584AAA BBB CCCC DD EE*{3600}DRB{4200}F125108405*SOMEWHER OVER THE RAINBOW*{4400}D125108405*NOT IN KANSAS*{5000}F125108405*LOST SOMEWHERE*{5400}125108405{6500}DRAW*CREDIT IMMUNITY PALACE*
run;
data want2;
infile fedwire truncover;
input @"{1100}" field1 :$5. @"{1110}" field2 :$12. @"{1120}" field3 :$34.;
run;
Was trying to use regular expression but can't figure that out either, I think this will work, but not sure how to keep field name prxparse('/(?<={\d\d\d\d})\w+/')
Any help would be appreciated. Thanks
... View more
